Stable Diffusion 3 Medium
Stable Diffusion 3 Medium is the latest and most advanced text-to-image AI model in our Stable Diffusion 3 series, comprising two billion parameters. It excels in photorealism, processes complex prompts, and generates clear text. The weights are available under an open non-commercial license. For commercial use, please contact us for licensing details.
Stable Diffusion XL
Get involved with the fastest growing open software project. Download and join other developers in creating incredible applications with Stable Diffusion XL as a foundation model.
SDXL Turbo
Discover the groundbreaking SDXL Turbo, the latest advancement from our research team. Built on the robust foundation of Stable Diffusion XL, this ultra-fast model transforms the way you interact with technology.
Read the Research Paper
Open, unbiased, and ready for the world
The next step in our mission to democratize AI and building a global foundation to activate humanity’s potential is to make our models as unbiased as possible. The introduction of our Japanese models is a clear step in that direction.
-
Japanese Stable Diffusion XL
A text-to-image model that emphasizes Japanese input to better understand Japanese culture. It comprehends Japanese expressions better than traditional English models, allowing for high-quality image generation that feels native.
Download Code
Learn More -
Japanese Stable VLM
Equipped with image captioning capabilities, this advanced Japanese image-language model can generate detailed Japanese descriptions for any image.
-
Japanese Stable Clip
A feature extraction model for zero-shot image classification and image search. Capable of classifying images without prior class information learning, it can also search for images based on any text.
Download Code
Learn More
Build with a Stability AI
Stability AI licenses offer flexibility for your generative AI needs by combining our range of state-of-the-art open models with self-hosting benefits.